A calculation method and system for the volatility distribution of organic aerosols in different particle size segments

Abstract

The present invention discloses a method and system for calculating the volatility distribution of organic aerosols in different particle size segments. The method comprises: correcting the volatility distribution data of particulate matter; measuring the chemical component data of particulate matter in different particle size segments; calculating the residence time of particulate matter in a heating tube; improving the multivariate kinetic volatility model based on the mass conversion process of particulate matter in the heating tube; performing sensitivity testing on the improved multivariate kinetic volatility model; adjusting the improved multivariate kinetic volatility model based on the test results, and inferring the volatility distribution of organic aerosols in different particle size segments. The system comprises a correction module, a chemical component measurement module, a residence time calculation module, a model improvement module, a sensitivity test module, and a volatility measurement module. By using the present invention, the error caused by factors such as completely volatile components can be removed, and the accuracy of inferring the volatility distribution of organic aerosols can be improved. The present invention can be widely applied to the technical field of volatility distribution calculation.

Description

Technical Field

[0001] The invention relates to the technical field of volatility distribution calculation, and in particular to a calculation method and system for volatility distribution of organic aerosols in different particle size segments. Background Art

[0002] Organic aerosol (OA) is an important component of atmospheric aerosol, which has an important impact on global climate change, atmospheric particulate pollution and human health. OA can be divided into primary organic aerosol (POA) and secondary organic aerosol (SOA) according to its source. Studies have shown that SOA has a greater mass contribution to OA in most atmospheric environments. However, people still lack sufficient understanding of the physical and chemical properties and sources of POA and SOA, making it difficult for existing atmospheric chemistry models to accurately simulate the evolution of OA in the atmosphere. The volatility of OA, as an important physical property, determines the gas-particle conversion process of SOA, which in turn affects the mass concentration, chemical composition and particle size distribution of OA.

[0003] At present, the study on OA volatility uses a volatility tandem analyzer (Volatility Tandem Differential Mobility Analyzer, VTDMA) combined with a multivariate kinetic volatility model to analyze the volatility distribution of OA in different particle size segments. The multivariate kinetic volatility model is a model used to simulate the mass flux of a single particle under different environmental and temperature conditions. The model can use forward simulation, that is, the information on the chemical composition of the particle and the volatility of each chemical component is known, combined with parameters such as particle size and ambient temperature, to simulate the mass change process of the particle under different temperature environments. It can also use reverse simulation, that is, the residual mass of the particle volatilized in different temperature environments is known, combined with the above information such as particle size, to infer the volatility distribution information of each component in the particle. However, when using a volatility tandem analyzer for measurement, the presence of completely volatile components (Compounds Volatile, CV) will lead to an overestimation of the residual mass (Mass Fraction Remaining, MFR), which will cause analysis deviations. In addition, the passing rate of the heating tube for particles of different particle sizes will also affect the analysis of OA volatility. In addition, when using the multivariate kinetic volatility model for simulation, the change in particle size caused by mass change is not taken into account, which leads to deviations in the simulation calculations. Summary of the invention

[0004] In order to solve the above technical problems, the purpose of the present invention is to provide a method and system for calculating the volatility distribution of organic aerosols in different particle size segments, which can remove errors caused by factors such as completely volatile components and improve the accuracy of estimating the volatility distribution of organic aerosols.

[0005] The first technical solution adopted by the present invention is: a method for calculating the volatility distribution of organic aerosols in different particle size segments, comprising the following steps:

[0006] Analyze and correct the particle volatility distribution data to obtain the corrected volatile residual mass fraction;

[0007] Measure the chemical composition data of particles in different particle size ranges based on aerosol chemical composition measurement equipment;

[0008] Calculate the residence time of particles inside the heating tube based on the sampling flow rate;

[0009] The multivariate kinetic volatility model is improved based on the mass conversion process of particles in the heating tube to obtain an improved multivariate kinetic volatility model.

[0010] Based on the particle volatility distribution data, the residence time, the chemical component data and the corrected volatile residual mass fraction, the improved multivariate kinetic volatility model is subjected to a sensitivity test to obtain an optimal mass retention coefficient and evaporation enthalpy;

[0011] The improved multivariate kinetic volatility model is adjusted based on the optimal mass retention coefficient and evaporation enthalpy, and the chemical composition data of the measured particulate matter and the volatile residual mass fraction of the measured particulate matter are used as inputs of the model to infer the volatility distribution of organic aerosols in different particle size segments of the measured particulate matter.

[0012] Furthermore, the step of analyzing and correcting the particle volatility distribution data to obtain a corrected volatile residual mass fraction specifically includes:

[0013] The volatilization reduction factor of the particles is measured by instruments;

[0014] The volatility reduction factor of the particles is corrected based on the TDMAfit algorithm to obtain the probability distribution function of the volatility reduction factor of particles with different particle sizes;

[0015] Calculating the number fraction of particles based on the probability distribution function of the volatilization reduction factor of particles of different particle sizes;

[0016] Measure the particle number concentration of non-volatile particles in the process of passing through the heating tube to obtain the heating tube passing rate;

[0017] Obtain the particle number concentration spectrum distribution through the instrument;

[0018] Calculating the number concentration ratio of completely volatile components based on the heating tube passing rate and the particle number concentration spectrum distribution;

[0019] The volatile residual mass fraction of the particles is calculated based on the number fraction of the particles, the volatilization reduction factor of the particles and the number concentration ratio of the completely volatile components to obtain a corrected volatile residual mass fraction.

[0020] Through this preferred step, the particle size change during the volatilization process of the particles, the completely volatile components and the particle size difference error factors of the heating tube pass rate are taken into consideration, so that the corrected volatile residual mass fraction can overcome these error factors.

[0021] Furthermore, the calculation expression of the corrected volatile residual mass fraction is as follows:

[0022]

[0023]

[0024]

[0025]

[0026] Among them, D p represents the specific measured particle size, T represents the temperature, MFR represents the volatile residue mass fraction, VFR represents the volatile residue volume fraction. Assuming that the particle density does not change before and after heating, the volatile residue volume fraction is equivalent to the volatile residue mass fraction; VSF represents the change in particle size after heating, D0 represents the particle size after the first electric mobility analyzer screening, Dp(T) represents the particle size after heating, and VSF represents the particle size after heating. i Indicates the fth particle size change value, NF i VSF represents the particle size change value of the particles i The number fraction, NF CV represents the number concentration ratio of completely volatile components, c represents the probability distribution function of the volatilization reduction factor of particles, N(D p ) represents the particle number concentration measured by the scanning electric mobility particle size spectrometer, that is, the particle number concentration before heating, N'(D p , T) represents the particle number concentration after heating measurement by the volatile tandem mobility analyzer, and η represents the heating tube pass rate.

[0027] Furthermore, the mass conversion process of the particles in the heating tube is calculated as follows:

[0028]

[0029]

[0030]

[0031]

[0032]

[0033]

[0034] Among them, D p (t) represents the particle size at time t, t0 represents the initial time, m p，i (t0) represents the mass of species f at the initial moment, ρ i represents the density of species f, D i represents the diffusion coefficient of each component of species f, f(Kn, α) represents the correction factor for discontinuous mass transfer, α represents the mass retention coefficient, Kn represents the Knudsen number, C i，g (t) represents the gas phase concentration of species f at time t, f i represents the mole fraction of species f, represents the saturated vapor pressure of species f at temperature T, σ s / a represents the surface tension of particles, M i represents the molar mass of species f, D p represents the specific measured particle size, R represents the ideal gas constant, T represents the temperature, N p (D p0 ) indicates the particle size D p0 The particle number concentration, λ i represents the molecular free path of species f in air, c i represents the average velocity of molecules f, represents the saturated vapor pressure of species f at temperature T, T ref Indicates temperature 298.15K, ΔH vap，i represents the enthalpy of vaporization.

[0035] Further, the residence time is calculated as follows:

[0036]

[0037] Where Q represents the sampling flow rate, L represents the length of the heating tube, S represents the cross-sectional area of ​​the heating tube, and T res Indicates the residence time of particles inside the heating tube.

[0038] Further, the step of performing a sensitivity test on the improved multivariate kinetic volatility model based on the particle volatility distribution data, the residence time, the chemical component data and the corrected volatile residual mass fraction to obtain the optimal mass residence coefficient and evaporation enthalpy specifically includes:

[0039] Using the particle volatility distribution data, the residence time, and the chemical component data as inputs of the improved multivariate kinetic volatility model, performing forward simulation to obtain a simulated volatile residual mass fraction;

[0040] taking the corrected volatile residual mass fraction as the actual volatile residual mass fraction, and calculating the least square residual based on the simulated volatile residual mass fraction;

[0041] The optimal mass retention factor and evaporation enthalpy are determined based on the least square residual.

[0042] The second technical solution adopted by the present invention is: a calculation system for the volatility distribution of organic aerosols in different particle size segments, comprising:

[0043] A correction module is used to analyze and correct the particle volatility distribution data to obtain the corrected volatile residual mass fraction;

[0044] Chemical composition measurement module, which measures the chemical composition data of particles of different particle size segments based on aerosol chemical composition measurement equipment;

[0045] A residence time calculation module calculates the residence time of particles inside the heating tube based on the sampling flow rate;

[0046] Model improvement module, which improves the multivariate kinetic volatility model based on the mass conversion process of particulate matter in the heating tube to obtain an improved multivariate kinetic volatility model;

[0047] A sensitivity testing module, which performs a sensitivity test on the improved multivariate kinetic volatility model based on the particle volatility distribution data, the residence time, the chemical component data and the corrected volatile residual mass fraction to obtain an optimal mass residence coefficient and evaporation enthalpy;

[0048] The volatility calculation module adjusts the improved multivariate kinetic volatility model based on the optimal mass retention coefficient and evaporation enthalpy, and uses the chemical composition data of the measured particulate matter and the volatile residual mass fraction of the measured particulate matter as inputs to the model to infer the volatility distribution of organic aerosols in different particle size segments of the measured particulate matter.

[0049] The beneficial effects of the method and system of the present invention are as follows: the present invention takes into account the particle size variation during the volatilization process of particulate matter, the completely volatile components and the particle size difference error factors of the heating tube pass rate; by analyzing the measurement data of the volatile tandem mobility analyzer and the scanning mobility particle size spectrometer, the volatile residual mass fraction is corrected in combination with the analysis data, so that the corrected volatile residual mass fraction can overcome these error factors; the multivariate kinetic volatility model is improved through the mass conversion process of the particulate matter in the heating tube, the multivariate kinetic volatility model is subjected to a sensitivity test through the corrected volatile residual mass fraction, and the improved multivariate kinetic volatility model is adjusted according to the obtained optimal mass retention coefficient and evaporation enthalpy, finally achieving the removal of errors caused by factors such as completely volatile components, improving the accuracy of inferring the volatile distribution of organic aerosols, and providing important support for the analysis of the causes of atmospheric pollution and the study of climate change. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is a flow chart of the steps of a method for calculating the volatility distribution of organic aerosols in different particle size segments of the present invention;

[0051] Figure 2 It is a structural diagram of a calculation system for the volatility distribution of organic aerosols in different particle size segments of the present invention;

[0052] Figure 3 It is a complete flow chart of a method for calculating the volatility distribution of organic aerosols in different particle size segments of the present invention;

[0053] Figure 4 It is a schematic diagram of a volatile tandem electric mobility analyzer for calculating a method for volatility distribution of organic aerosols in different particle size segments of the present invention;

[0054] Figure 5 is a graph showing the average particle size distribution of a specific embodiment of the present invention;

[0055] Figure 6 is a schematic diagram of the heating tube passing rate of a specific embodiment of the present invention;

[0056] Figure 7 is a schematic diagram of the number concentration ratio of a specific embodiment of the present invention;

[0057] Figure 8 is a schematic diagram of the volatile residual mass fraction of a specific embodiment of the present invention;

[0058] Fig. 9 is a volatility distribution diagram of a specific embodiment of the present invention;

[0059] Fig.10 It is a comparison diagram of the volatility distribution of organic aerosols in different particle size segments during the day and at night according to a specific embodiment of the present invention. DETAILED DESCRIPTION

[0060] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. The step numbers in the following embodiments are only provided for the convenience of explanation and description, and the order between the steps is not limited in any way. The execution order of each step in the embodiment can be adaptively adjusted according to the understanding of those skilled in the art.

[0061] Reference Figure 1 and Figure 3 The present invention provides a method for calculating the volatility distribution of organic aerosols in different particle size segments, the method comprising the following steps:

[0062] S1. Analyze and correct the particle volatility distribution data to obtain the corrected volatile residual mass fraction;

[0063] S1.1. Measure the volatility reduction factor of particles using a volatile tandem mobility analyzer;

[0064] Specifically, refer to Figure 4 The volatile tandem mobility analyzer (VTDMA) consists of two mobility analyzers (DMAs), a heating tube (Heater) and a particle counter (CPC). The particles first pass through the first mobility analyzer (DMA1) for particle size screening, and the particles with a specific particle size (D0) enter the heating tube. The heating tube is temperature controlled and stabilized at a specific temperature (T=T1, T2, T3,...). After passing through the heating tube, the particles will volatilize due to heat, resulting in changes in particle size. The heated particles enter the second mobility analyzer (DMA2). DMA2 uses a scanning mode. By changing the voltage applied to DMA2, the particle size of the particles passing through DMA2 is controlled. Combined with the particle counter, the particle size distribution of the particles after thermal volatilization can be known. The volatility reduction factor is defined as the particle size change VSF (Volatility Shrink Factor, VSF) of the particles after heating, and its calculation expression is as follows:

[0065]

[0066] Wherein, T represents temperature, D0 represents the particle size after screening by the first mobility analyzer, and Dp(T) represents the particle size after heating.

[0067] S1.2. Based on the TDMAfit algorithm, the volatility reduction factor of the particles is corrected to eliminate the deviation caused by factors such as the DMA slit, and the volatility reduction factor probability distribution function (VolatilityShrink Factor Probability Density Function, VSF-PDF) of particles of different particle sizes is obtained;

[0068] S1.3, calculating the number fraction of particles based on the probability distribution function of the volatilization reduction factor of particles of different particle sizes;

[0069] Specifically, the number fraction of the particles is expressed as follows:

[0070]

[0071] Among them, NF i VSF represents the particle size change value of the particles i The number fraction at that time, c represents the probability distribution function of the volatilization shrinkage factor of the particles, D p Indicates specific measurement particle size, VSF i Represents the i-th particle size change value.

[0072] S1.4. Measure the particle number concentration of non-volatile particles in the process of passing through the heating tube to obtain the heating tube passing rate;

[0073] Specifically, the mass fraction remaining (MFR) refers to the ratio of the mass of the particles after volatilization to the mass before volatilization. MFR is affected by the completely volatile components and the heating tube pass rate. In this embodiment, non-volatile sodium chloride (NaCl) particles are used to calibrate the heating tube pass rate. NaCl particles of different particle sizes are passed into heating tubes of different temperatures. By testing the particle number concentration before and after entering and passing through, the heating tube pass rate η (D p , T).

[0074] S1.5. Obtain the particle number concentration spectrum distribution through the instrument;

[0075] Specifically, the particle number concentration spectrum distribution can be directly measured and obtained by a scanning electric mobility particle size spectrometer.

[0076] S1.6. Calculating the number concentration ratio of completely volatile components based on the heating tube passing rate and the particle number concentration spectrum distribution;

[0077] Specifically, the number concentration ratio of the completely volatile components is calculated as follows:

[0078]

[0079] Among them, NF CV represents the number concentration ratio of completely volatile components, N(D p ) represents the particle number concentration measured by the scanning electric mobility particle size spectrometer, that is, the particle number concentration before heating, N′(D p , T) represents the particle number concentration after heating measurement by the volatile tandem mobility analyzer, and η represents the heating tube pass rate.

[0080] S1.7. Calculate the volatile residual mass fraction of the particles based on the number fraction of the particles, the volatilization reduction factor of the particles and the number concentration ratio of the completely volatile components to obtain a corrected volatile residual mass fraction.

[0081] Specifically, the corrected volatile residual mass fraction is calculated by the following expression:

[0082]

[0083] Among them, MFR represents the volatile residual mass fraction, VFR represents the volatile residual volume fraction, and assuming that the particle density does not change before and after heating, the volatile residual volume fraction is equivalent to the volatile residual mass fraction.

[0084] S2. Measure the chemical composition data of particles in different particle size ranges based on aerosol chemical composition measurement equipment;

[0085] Specifically, the chemical components of particles in different particle size segments include organic aerosol, ammonium sulfate, ammonium nitrate and black carbon particles. According to the volatility basis set (VBS), the organic aerosol is divided into the following categories according to its saturated vapor pressure (C*(T ref ), T ref =298.15K), divided into three types of organic aerosols: Extremely Low Volatility Organic Aerosol (ELVOA, C * =10 -5 ug m -3 ), Low Volatility Organic Aerosol (LVOA, C * =10 -2 ug m -3 ) and semi-volatile organic aerosol (SVOA, C * =10 1 ug m -3 ).

[0086] The chemical composition data of particles of different particle size segments also include the parameters of each chemical component, including the diffusion coefficient D of each component. i , surface tension σ s / a , molar mass M i , density ρ i , enthalpy of vaporization ΔH vap，i and mass retention coefficient α, the parameters of each chemical component refer to Table 1.

[0087] Table 1 Parameters of each chemical component

[0088]

[0089] The α of each component and the ΔH of ELVOA, LVOA and SVOA vap，i This needs to be determined through sensitivity testing.

[0090] S3, calculating the residence time of the particles inside the heating tube based on the sampling flow rate;

[0091] Specifically, the residence time (T res ) is an important parameter for simulating the volatilization process of particulate matter, and its calculation expression is as follows:

[0092]

[0093] Where Q represents the sampling flow rate, L represents the length of the heating tube, S represents the cross-sectional area of ​​the heating tube, and T res Indicates the residence time of particles inside the heating tube.

[0094] S4, improving the multivariate kinetic volatility model based on the mass conversion process of the particles in the heating tube to obtain an improved multivariate kinetic volatility model;

[0095] Specifically, the mass conversion process of the particles in the heating tube is calculated as follows:

[0096]

[0097]

[0098]

[0099]

[0100]

[0101]

[0102] Among them, D p (t) represents the particle size at time t, t0 represents the initial time, mp，i (t0) represents the mass of species f at the initial moment, ρ i represents the density of species f, D i represents the diffusion coefficient of each component of species f, f(Kn, α) represents the correction factor for discontinuous mass transfer, α represents the mass retention coefficient, Kn represents the Knudsen number, C i，g (t) represents the gas phase concentration of species f at time t, f i represents the mole fraction of species f, represents the saturated vapor pressure of species f at temperature T, σ s / a represents the surface tension of particles, M i represents the molar mass of species i, D p represents the specific measured particle size, R represents the ideal gas constant, T represents the temperature, N p (D p0 ) indicates the particle size D p0 The particle number concentration, λ i represents the molecular free path of species f in air, c i represents the average velocity of molecules f, represents the saturated vapor pressure of species f at temperature T, T ref Indicates temperature 298.15K, ΔH vap，i represents the enthalpy of vaporization.

[0103] The mass conversion process of the particles in the heating tube is applied to the multivariate kinetic volatility model, so that the particle size of the particles at time t in the improved multivariate kinetic volatility model can also take the particle size change during the heating process into account.

[0104] S5. Based on the particle volatility distribution data, the residence time, the chemical composition data and the corrected volatile residual mass fraction, the improved multivariate kinetic volatility model is subjected to a sensitivity test to obtain an optimal mass retention coefficient and evaporation enthalpy;

[0105] Specifically, before formally using the multivariate dynamic volatility model to estimate the volatility distribution of organic aerosols, it is necessary to first use a sensitivity test to determine the evaporation enthalpy (ΔH vap，i ) and mass retention coefficient (α).

[0106] S5.1. Using the particle volatility distribution data, the residence time, and the chemical component data as inputs of the improved multivariate kinetic volatility model, performing forward simulation to obtain a simulated volatile residual mass fraction;

[0107] Specifically, different combinations of enthalpy of evaporation ΔH are used. vap，i And the mass retention coefficient α is simulated, and the simulation expression of evaporation enthalpy is as follows:

[0108]

[0109] Where ΔH vap，i represents the evaporation enthalpy, a and b represent the linear fitting parameters, T ref Indicates temperature 298.15K.

[0110] The mass retention coefficient α does not need to be set as a linear fitting parameter like the evaporation enthalpy. The mass retention coefficient α can be set as a constant parameter.

[0111] The simulated evaporation enthalpy ΔH vap，i and mass retention coefficient α, and then combined with the particle volatility distribution data, the residence time, and the chemical component data as inputs of the improved multivariate kinetic volatility model, the volatile residual mass fraction under the simulated evaporation enthalpy and mass retention coefficient can be obtained.

[0112] When calculating the simulated volatile residual mass fraction, it is necessary to pay attention to the constraints of the model. The constraints are expressed as follows:

[0113] ∑f i，inorganics +∑f i，organics =1

[0114] Among them, f i，inorganic s represents the mass ratio of inorganic component i, f i，organics Indicates the mass proportion of the organic component f.

[0115] S5.2, taking the corrected volatile residual mass fraction as the actual volatile residual mass fraction, and calculating the least square residual based on the simulated volatile residual mass fraction;

[0116] Specifically, the calculation expression of the least square residual is as follows:

[0117]

[0118] Among them, SSR represents the least square residual, MFR model (T i , Dp) represents the simulated volatile residue mass fraction, MFR measured (T i , Dp) represents the actual volatile residual mass fraction.

[0119] S5.3. Determine the optimal mass retention coefficient and evaporation enthalpy based on the least square residual.

[0120] Specifically, find the minimum square residual value and consider the ΔH corresponding to the minimum square residual value vapAnd α is the optimal evaporation enthalpy and mass retention coefficient, at which time the simulated volatile residual mass fraction is closest to the actual volatile residual mass fraction. In addition, in order to improve the simulation efficiency, the actual volatile residual mass fraction of this scheme adopts the average value of the corrected volatile residual mass fraction.

[0121] S6. Based on the optimal mass retention coefficient and evaporation enthalpy, the improved multivariate kinetic volatility model is adjusted, and the chemical composition data of the measured particulate matter and the volatile residual mass fraction of the measured particulate matter are used as inputs of the model to infer the volatility distribution of organic aerosols in different particle size segments of the measured particulate matter.

[0122] Specifically, after determining the optimal mass retention coefficient and evaporation enthalpy, the linear fitting parameters of the evaporation enthalpy and the constant parameters of the mass retention coefficient are introduced into the improved multivariate kinetic volatility model. The chemical composition data of the measured particulate matter and the volatile residual mass fraction of the measured particulate matter are used as the input of the model. At this time, the model performs inverse operations and outputs the volatility distribution of organic aerosols in different particle size segments of the measured particulate matter.

[0123] The specific embodiment of the present invention carries out an outdoor observation experiment on the Heshan Environmental Monitoring Super Station in Jiangmen, Guangdong. The online monitoring instrument and its monitoring parameters are shown in Table 2.

[0124] Table 2 Online monitoring instruments and their monitoring parameters

[0125]

[0126] Among them, the temperature in the heating tube of the volatile tandem mobility analyzer (VTDMA) was set to 25°C, 100°C, 150°C, 200°C and 250°C, and the carbon black time-of-flight aerosol mass spectrometry (SP ToF-AMS) measured the chemical components of particles with Stokes particle sizes ranging from 31.25nm to 781.25nm.

[0127] Reference Figure 5 Due to the low mass concentration of particles below 100 nm, it is difficult for the carbon black time-of-flight aerosol mass spectrometer to give accurate concentration information. Here, a bimodal log-normal distribution fitting is performed on the hourly average data of the carbon black time-of-flight aerosol mass spectrometer to reduce the uncertainty of the instrument in the small particle size segment due to the low mass concentration of particles.

[0128] The NH4 + 、SO4 2- and NO3 - The number of ions is converted into the number of inorganic salt components, and the calculation expression is as follows:

[0129]

[0130]

[0131]

[0132]

[0133]

[0134] in, Indicates the amount of ammonium nitrate, represents the amount of nitrate ions, Indicates the amount of sulfuric acid, represents the amount of sulfate ions, Indicates the number of ammonium ions, Indicates the amount of ammonium bisulfate, Indicates the amount of ammonium sulfate, Indicates the amount of nitric acid.

[0135] Reference Figure 6 As a preferred embodiment, before and after the observation in step S1.4, the particle passing rate η (D p , T), where the temperature setting of the heating tube should be consistent with the temperature setting during observation (25℃, 100℃, 150℃, 200℃ and 250℃). It can be found that the passing rate of small-sized particles is lower than that of large-sized particles, and the higher the temperature, the lower the passing rate of particles.

[0136] Reference Figure 7 As a preferred embodiment, when determining the passing rate (η(D p After that, the number concentration (N(D) of particles with sizes of 30, 50, 80, 100, 150 and 200 nm was measured by a scanning mobility particle size spectrometer. p )), and substitute it into the formula in step S1.6 to calculate the number concentration ratio of the completely volatile components (NF CV (D p , T)). Taking the average concentration percentage data during the observation period as an example ( Figure 7 ), the number concentration ratio of the completely volatile components of the particles increases with the increase of temperature, which means that the completely volatile component of the particles is higher when the temperature is higher. At the same time, the number concentration ratio of the completely volatile components of the small-sized particles is also higher than that of the large-sized particles.

[0137] Reference Figure 8 As a preferred embodiment, the NF of particles with different particle sizes at different heating temperatures is obtained. cV, the corrected volatile residual mass fraction can be calculated by the probability distribution function of the volatility reduction factor of particles of different particle sizes measured by the volatile tandem mobility analyzer using the formula in step S1.7. The average data of the corrected volatile residual mass fraction during the observation period is taken as an example ( Figure 8 ), in general, the corrected volatile residual mass fraction decreases with increasing temperature, which means that the residual volume of particles is smaller at higher temperatures. In addition, the corrected volatile residual mass fraction of large-size particles is higher than that of small-size particles at high temperatures, and the rate of decline is also slower than that of small-size particles, indicating that large-size particles are generally less volatile.

[0138] Referring to Table 3, as a preferred embodiment, the average value of the volatile residue mass fraction corrected during the observation period is used for sensitivity testing. In the test, the linear fitting parameter a is taken as 0, 4, 8, 12, and the linear fitting parameter b is taken as 50, 80, 100, 150, 200. Under this combination, the evaporation enthalpy ΔH vap，i There are 20 values ​​in total, and the mass retention coefficient α is 0.01, 0.09, 0.1, 0.5, 0.7, 0.9, and 1. It can be found from Table 3 that the least square residual SSR is the smallest when a=0, b=100, and α=0.09, indicating that the simulated volatile residual mass fraction MFR is model (T i , Dp) and the actual volatile residual mass fraction MFR measured (T i , Dp) is closest, so the parameter combination of a=0, b=100, α=0.09 is used to estimate the volatility of organic aerosol.

[0139] Table 3 Different ΔH vap，i And the SSR value under α combination

[0140] α＝0.01 α＝0.09 α＝0.1 α＝0.5 α＝0.7 α＝0.9 α＝1 a=0,b=50 4.147 0.929 0.841 0.087 0.056 0.052 0.054 a=0,b=80 0.299 0.038 0.037 0.022 0.019 0.018 0.018 a=0,b=100 0.056 0.016 0.018 0.107 0.135 0.157 0.169 a=0,b=150 0.504 0.657 0.671 1.040 1.158 1.245 1.281 a=0,b=200 1.305 1.378 1.382 1.907 2.031 2.179 2.434 a=4,b=50 1.634 0.205 0.176 0.030 0.027 0.024 0.022 a=4,b=80 0.121 0.141 0.151 0.335 0.366 0.395 0.407 a=4,b=100 0.272 0.480 0.497 0.643 0.653 0.680 0.697 a=4,b=150 0.926 1.377 1.381 1.440 1.454 1.472 1.484 a=4,b=200 1.305 2.457 2.514 3.585 3.663 3.690 3.697 a=8,b=50 0.588 0.216 0.222 0.369 0.409 0.432 0.442 a=8,b=80 0.429 0.636 0.638 0.849 0.941 1.005 1.037 a=8,b=100 0.581 1.003 1.037 1.433 1.450 1.460 1.464 a=8,b=150 1.305 1.427 1.456 2.630 2.862 3.045 3.118 a=8,b=200 2.636 3.601 3.605 3.669 3.685 3.696 3.701 a=12,b=50 0.646 0.640 0.642 0.922 1.002 1.081 1.116 a=12,b=80 0.790 1.368 1.375 1.439 1.454 1.463 1.466 a=12,b=100 1.303 1.378 1.382 1.562 1.922 2.004 2.052 a=12,b=150 1.744 3.270 3.340 3.668 3.685 3.696 3.700 a=12,b=200 3.527 3.601 3.605 3.669 3.685 3.696 3.701

[0141] Reference Fig. 9 As a preferred embodiment, after obtaining ΔH through sensitivity analysis vap After the parameters of α and α are combined, they are substituted into the improved multivariate dynamic volatility model, combined with the measured particle chemical composition, MFR measured (T i , Dp) and other data and parameters are used to estimate the volatility distribution of organic aerosols in different particle size segments in real-time measurement. Fig. 9The average distribution of organic aerosol volatility in different particle size segments during the observation period calculated by the model is shown. It can be found that the proportion of semi-volatile organic aerosol SVOA in the small particle size segment is relatively high, while the proportion of extremely low volatile organic aerosol ELVOA is relatively low. As the particle size increases, the proportion of semi-volatile organic aerosol SVOA gradually decreases, while the proportion of extremely low volatile organic aerosol ELVOA gradually increases. This shows that the volatility of organic aerosol decreases with the increase of particle size, which may be due to the fact that large-size particles are more aged due to the atmospheric oxidation process.

[0142] Reference Fig.10 As a preferred embodiment, the volatility distribution of organic aerosols in different particle size segments during the day (8:00-16:00) and night (20:00-4:00) was analyzed. It can be found that the volatility of organic aerosols in various particle size segments during the day is higher than that at night. This may be because the photochemical reaction is more intense during the day, causing gas-phase volatile organic compounds (VOCs) to be transferred to particulate matter through gas-particle distribution and other effects, and these "fresh" organic components attached to the particulate matter are often more volatile.

[0143] Reference Figure 2 The present invention provides a calculation system for the volatility distribution of organic aerosols in different particle size segments, comprising:

[0144] A correction module is used to analyze and correct the particle volatility distribution data to obtain the corrected volatile residual mass fraction;

[0145] Chemical composition measurement module, which measures the chemical composition data of particles of different particle size segments based on aerosol chemical composition measurement equipment;

[0146] A residence time calculation module calculates the residence time of particles inside the heating tube based on the sampling flow rate;

[0147] Model improvement module, which improves the multivariate kinetic volatility model based on the mass conversion process of particulate matter in the heating tube to obtain an improved multivariate kinetic volatility model;

[0148] A sensitivity testing module, which performs a sensitivity test on the improved multivariate kinetic volatility model based on the particle volatility distribution data, the residence time, the chemical component data and the corrected volatile residual mass fraction to obtain an optimal mass residence coefficient and evaporation enthalpy;

[0149] The volatility calculation module adjusts the improved multivariate kinetic volatility model based on the optimal mass retention coefficient and evaporation enthalpy, and uses the chemical composition data of the measured particulate matter and the volatile residual mass fraction of the measured particulate matter as inputs to the model to infer the volatility distribution of organic aerosols in different particle size segments of the measured particulate matter.

[0150] The contents of the above method embodiments are all applicable to the present system embodiments. The functions specifically implemented by the present system embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0151] The above is a specific description of the preferred implementation of the present invention, but the invention is not limited to the embodiments. Those skilled in the art may make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of this application.

Claims

1. A method for calculating the volatility distribution of organic aerosols in different particle size segments, characterized in that: The following steps are involved: Analyze and correct the particle volatility distribution data to obtain the corrected volatile residual mass fraction; Measure the chemical composition data of particles in different particle size ranges based on aerosol chemical composition measurement equipment; Calculate the residence time of particles inside the heating tube based on the sampling flow rate; The multivariate kinetic volatility model is improved based on the mass conversion process of particles in the heating tube to obtain an improved multivariate kinetic volatility model. Based on the particle volatility distribution data, the residence time, the chemical component data and the corrected volatile residual mass fraction, the improved multivariate kinetic volatility model is subjected to a sensitivity test to obtain an optimal mass retention coefficient and evaporation enthalpy; The improved multivariate kinetic volatility model is adjusted based on the optimal mass retention coefficient and evaporation enthalpy, and the chemical component data of the measured particulate matter and the volatile residual mass fraction of the measured particulate matter are used as inputs of the model to infer the volatility distribution of organic aerosols in different particle size segments of the measured particulate matter; The step of analyzing and correcting the particle volatility distribution data to obtain a corrected volatile residual mass fraction specifically includes: The volatilization reduction factor of the particles is measured by instruments; The volatility reduction factor of the particles is corrected based on the TDMAfit algorithm to obtain the probability distribution function of the volatility reduction factor of particles with different particle sizes; Calculating the number fraction of particles based on the probability distribution function of the volatilization reduction factor of particles of different particle sizes; Measure the particle number concentration of non-volatile particles in the process of passing through the heating tube to obtain the heating tube passing rate; Obtain the particle number concentration spectrum distribution through the instrument; Calculating the number concentration ratio of completely volatile components based on the heating tube passing rate and the particle number concentration spectrum distribution; The volatile residual mass fraction of the particles is calculated based on the number fraction of the particles, the volatilization reduction factor of the particles and the number concentration ratio of the completely volatile components to obtain a corrected volatile residual mass fraction.

2. The method for calculating the volatility distribution of organic aerosols in different particle size segments according to claim 1, characterized in that: The calculation expression of the corrected volatile residual mass fraction is as follows: Among them, D p represents the preset measured particle size, T represents temperature, MFR represents volatile residual mass fraction, VFR represents volatile residual volume fraction. Assuming that the particle density does not change before and after heating, the volatile residual volume fraction is equivalent to the volatile residual mass fraction; VSF represents the change in particle size after heating, D0 represents the particle size after screening by the first mobility analyzer, Dp(T) represents the particle size after heating, and VSF represents the particle size after heating. i Indicates the change value of the i-th particle size, NF i VSF represents the particle size change value of the particles i The number fraction, NF CV represents the number concentration ratio of completely volatile components, c represents the probability distribution function of the volatilization reduction factor of particles, N(D p ) represents the particle number concentration measured by the scanning electric mobility particle size spectrometer, that is, the particle number concentration before heating, N'(D p ,T) represents the particle number concentration after heating measurement by the volatile tandem mobility analyzer, and η represents the heating tube pass rate.

3. The method for calculating the volatility distribution of organic aerosols in different particle size segments according to claim 1, characterized in that: The mass conversion process of the particles in the heating tube is calculated as follows: Among them, D p (t) represents the particle size at time t, t0 represents the initial time, m p,i (t0) represents the mass of species i at the initial moment, ρ i represents the density of species i, D i represents the diffusion coefficient of each component of species i, f(Kn,α) represents the correction factor for discontinuous mass transfer, α represents the mass retention coefficient, Kn represents the Knudsen number, C i,g (t) represents the gas phase concentration of species i at time t, f i represents the mole fraction of species i, represents the saturated vapor pressure of species i at temperature T, σ s / a represents the surface tension of particles, M i represents the molar mass of species i, D p represents the preset measurement particle size, R represents the ideal gas constant, T represents the temperature, N p (D p0 ) indicates the particle size D p0 The particle number concentration, λ i represents the molecular free path of species i in air, c i represents the average rate of molecule i, represents the saturated vapor pressure of species i at temperature T, T ref Indicates temperature 298.15K, ΔH vap,i represents the enthalpy of vaporization.

4. The method for calculating the volatility distribution of organic aerosols in different particle size segments according to claim 1, characterized in that: The residence time is calculated by the following expression: Where Q represents the sampling flow rate, L represents the length of the heating tube, S represents the cross-sectional area of ​​the heating tube, and T res Indicates the residence time of particles inside the heating tube.

5. The method for calculating the volatility distribution of organic aerosols in different particle size segments according to claim 1, characterized in that: The step of performing a sensitivity test on the improved multivariate kinetic volatility model based on the particle volatility distribution data, the residence time, the chemical component data and the corrected volatile residual mass fraction to obtain the optimal mass residence coefficient and evaporation enthalpy specifically includes: Using the particle volatility distribution data, the residence time, and the chemical component data as inputs of the improved multivariate kinetic volatility model, performing forward simulation to obtain a simulated volatile residual mass fraction; taking the corrected volatile residual mass fraction as the actual volatile residual mass fraction, and calculating the least square residual based on the simulated volatile residual mass fraction; The optimal mass retention factor and evaporation enthalpy are determined based on the least square residual.

6. A calculation system for the volatility distribution of organic aerosols in different particle size segments, characterized in that: A method for calculating the volatility distribution of organic aerosols in different particle size segments as claimed in claim 1, comprising: A correction module is used to analyze and correct the particle volatility distribution data to obtain the corrected volatile residual mass fraction; Chemical composition measurement module, which measures the chemical composition data of particles of different particle size segments based on aerosol chemical composition measurement equipment; A residence time calculation module calculates the residence time of particles inside the heating tube based on the sampling flow rate; Model improvement module, which improves the multivariate kinetic volatility model based on the mass conversion process of particulate matter in the heating tube to obtain an improved multivariate kinetic volatility model; A sensitivity testing module, which performs a sensitivity test on the improved multivariate kinetic volatility model based on the particle volatility distribution data, the residence time, the chemical component data and the corrected volatile residual mass fraction to obtain an optimal mass residence coefficient and evaporation enthalpy; The volatility calculation module adjusts the improved multivariate kinetic volatility model based on the optimal mass retention coefficient and evaporation enthalpy, and uses the chemical composition data of the measured particulate matter and the volatile residual mass fraction of the measured particulate matter as inputs to the model to infer the volatility distribution of organic aerosols in different particle size segments of the measured particulate matter.

Images